| 25 Jan 2025 |
 7c6f434c | Well, trying to get a company to do something tends to leave you with the ball out of your court! | 16:48:54 |
| 7c6f434c | But yeah, maybe you also need Graham to react, which increases the amount of not-yours courts in the game | 16:49:43 |
 adamcstephens | In reply to @winter:catgirl.cloud
(#2 is true, but a tad more complicated than that :)) Can you share more detail on what the blocker is? | 17:15:08 |
| adamcstephens | How much capacity are we talking about? Could be useful to know in this context | 17:27:12 |
 cole-h | In reply to @7c6f434c:nitro.chat
But yeah, maybe you also need Graham to react, which increases the amount of not-yours courts in the game AFAIK, both Graham and I are no longer in the critical path | 22:14:30 |
| cole-h | And I've just now deleted the ofborg cloudamqp instance (after double checking it was empty) | 22:26:41 |
| cole-h | cc @janne.hess:helsinki-systems.de-- may want to delete your shovelout user (I imagine that was yours at least), since the instance it's trying to shovel from has been deleted | 22:27:35 |
 hexa | 4 M1 Macs that were previously used for hydra.nixos.org 😄 | 23:07:47 |
| 26 Jan 2025 |
| adamcstephens | better than nothing :) | 00:49:30 |
| 27 Jan 2025 |
| hexa | 
Download image.png | 00:23:24 |
| hexa | meh. | 00:23:35 |
| adamcstephens | In reply to @hexa:lossy.network
sent an image. Is that page publicly available? | 01:08:58 |
| hexa | no | 01:09:12 |
| hexa | that is the rabbitmq mgmt dashboard | 01:09:25 |
| hexa | hence me posting these screenshots | 01:09:31 |
| adamcstephens | Can we hook up the infra Prometheus to expose data? | 01:09:57 |
| hexa | https://termbin.com/1qr4 | 02:39:06 |
| hexa | those are the metrics it exports 🤷♂️ | 02:39:16 |
| hexa | no per-queue metrics | 02:39:41 |
| hexa | https://www.rabbitmq.com/docs/monitoring#queue-metrics | 02:40:13 |
| hexa | * https://www.rabbitmq.com/docs/prometheus#metric-aggregation | 02:42:52 |
| hexa | maybe we just want the fat mode | 02:42:55 |
| hexa | yep, here they are | 02:44:13 |
| hexa | https://termbin.com/mlqq | 02:44:46 |
| hexa | https://www.rabbitmq.com/docs/prometheus#per-object-endpoint | 02:49:20 |
| hexa | oh and we get a 15% discount at oakhost, should we want to rent anything 🤔 | 02:54:37 |
| adamcstephens | So many erlang VM metrics :) | 02:54:45 |
| hexa | but the first step would probably to split the two m1 macs in the US between ofborg and hydra | 02:54:52 |
| hexa | * but the first step would probably to split the four m1 macs in the US between ofborg and hydra | 02:54:59 |
| hexa | because I'm not super happy about hydra's darwin performance atm either | 02:55:13 |
